1.Field of Invention
The present invention relates to cleaning analytical instruments such as Scanning Electron Microscopes (SEM), Scanning Electron Microprobes, Transmission Electron Microscopes (TEM) and other charge particle beam instruments that are subject to contamination problems from hydrocarbons. In particular it is a novel method and apparatus for cleaning the specimen chamber, specimen stage, and specimen in-situ inside the vacuum system of these instruments with oxygen radicals that uses air passed through a glow discharge as an oxygen radical source. The oxygen radicals are used to oxidize the hydrocarbons and convert them to easily pumped gases. The method and apparatus can be added to the analytical instrument with no change to its analytical purpose or design.
2. Description of Prior Art
Electron microscopy is used to detect, measure, and analyze constituents present in very small areas of materials. Contaminants adsorbed on the surface or surface films interacting with the incident electron probe beam can distort the results. Deposits created by the interaction of the probe beam with the surface specimen also may interfere with the probe bean or emitted electrons and x-rays and thus adversely affect accurate analysis. Deposits also add uncertainty to SEM measured line widths for semiconductor device critical dimension metrology.
Another problem is the condensation of pump oils on the windows of the x-ray and electron detectors distorting results. The most serious problem of this type is the absorption of low-energy x-rays from Be, C, N, O and F by oil films which can prevent measurement of these elements by X-ray emission spectroscopy.
Contaminants typically are introduced by one of four ways including the specimen, the specimen stage, carried into the chamber by the evacuation system, or are present on the internal components of the instrument. Contaminants introduced from the evacuation system can be reduced by trapping, by purging, or by using cleaner pumps. Once present inside the chamber these contaminants reside on the chamber surfaces, and can be removed only slowly and with low efficiency by the high vacuum pump.
Inorganic specimens (metals, ceramics, semiconductors, etc.) may carry contaminants into the chamber. These may be part of the specimen, residues from sample preparation techniques or be caused improper sample handling or storage techniques. In addition, clean surfaces will accumulate a surface film of hydrocarbon scum if left exposed to ordinary room air for any length of time. The sources of these hydrocarbons are most any living thing, organic object, or other source of hydrocarbon vapors in the vicinity. While the part of the films created in these processes dissipate under vacuum conditions, a small amount generally remains on surfaces and is sufficient to cause problems when the specimen is subsequently examined in the analytical instruments listed.
These residues are widely distributed and generally are at low concentrations on the various surfaces. Some of the contaminant molecules become mobile in the vacuum environment. At high vacuum the mean free path of molecules once vaporized is comparable to or longer than the dimensions of the vacuum chamber of these instruments. The contaminants move in the vapor phase from surface to surface in the vacuum environment and are attracted to any focused electron probe beam, forming deposits through an ionization and deposition process. Since these contaminants can travel large distances within the vacuum chamber and over the surface of a specimen, it is important to remove or immobilize these species as much as possible prior to an analysis without disturbing the microstructure of the specimen.
Ronald Vane, Dba XEI Scientific, has sold a nitrogen purge system for cleaning SEM chambers since 1991. Operating at a pressure of approximately 1 Torr in the chamber, this system uses viscous flow vacuum dynamics to carry contaminants from the chamber to the roughing pumps. This system is operated every night and needs at least 40 hours a week of operation to keep the chamber clean. It is not fast, it does not reactively clean, and cannot be used where 24 hr instrument availability is needed for the electron microscope. Another problem for the purge technique is the changing design of electron microscope vacuum systems. The latest design pumping systems use turbo molecular pump without a valve between the chamber and the pump. To vent the chamber the turbo pump is stopped and gas admitted to the chamber. During the pump down cycle, roughing takes place through the turbo molecular pump while it is accelerating. Any leak of gas into the chamber into the chamber during rough will result in an overheated turbo pump. Thus a continuous purge is not possible for this type of vacuum system.
It has been well documented that low temperature (&lt;500.degree. C.) plasmas of various ionized gases can be used to reactively etch/ash organic materials found on the surface of materials. As "glow-discharge cleaning" it has been used by the high energy physics community to condition the interiors of large vacuum vessels. Named "plasma etch" or "plasma ashing", it has been used in the industrial community to clean and etch semiconductor wafers and other bulk materials for many years. In the microscopy community RF or DC plasma, dry-ashing devices are sold by several vendors to clean electron microscope specimens prior to analysis. In this procedure, typically the material is placed in an RF cavity or a DC cavity with a flowing reactive gas. The nature of the gas selected is chosen based upon the desired effect. Argon, nitrogen, air, oxygen or other gas mixtures are commonly used, and gases (BCl.sub.3, CF.sub.4) may be used to tailor the reaction.
Most of the current literature and recent patents on glow-discharge cleaning and plasma etch is concerned with the use of these processes in semiconductor production. For these processes plasma uniformity, anisotropic etching, and other highly controlled properties are important. The geometry of these systems is very carefully designed for uniform results. A variety of gases can be used for etching and cleaning. Gases such as Hydrogen, Argon, Nitrogen, Oxygen, CF.sub.4 and gas mixtures such as air and argon/oxygen have successfully been used for glow-discharge cleaning and plasma etching. Depending on the process the importance of ion sputtering and reactive ion etching varies, but in most of processes the neutral free radicals are the most important reactive species in the plasma. The free radicals, because they are neutral, are able to leave the electric fields of the excitation region and travel throughout the chamber by convection.
For the cleaning and removal of hydrocarbons the reaction with oxygen radicals to produce CO, CO.sub.2 and H.sub.2 O is the most important. These reaction products are quickly removed as gases from the vacuum system. These reactions are the dominant reactions in glow discharge cleaning methods using oxygen as a reactant gas. The glow discharge is used to produce oxygen ions that are then transformed into oxygen radicals by subsequent reactions. The oxygen ions are not needed as the reactive species for hydrocarbons. In the absence of nitrogen ions or other reactive species that destroy O radicals, O radicals are long lived and have the ability to do isotropic cleaning on all surfaces in the chamber. CF.sub.4 or other fluorine containing gases are sometimes added to oxygen containing plasmas to speed the oxidation of hydrocarbons by performing hydrid extraction on the base molecules to make them more susceptible to oxygen attack. This mechanism is important in the oxidation and removal of polymers such as photoresist.
Electron Beam Lithography systems scan an electron beam over organic resist materials which spew contaminants throughout the vacuum system and electron optics. The deposition of contaminants on the optics elements such as apertures, deflectors and lenses degrades performance and precision drawing ability of these instruments. Glow-discharge and plasma cleaning devices and cleaning methods for electron optics are described in U.S. Pat. Nos. 5,312,519 (Sakai et al.), 5,539,211 (Ohtoshi et al.) and 4,665,315 (Bacchetti et al.). These patents do not address the solving of the contamination problem on specimens within analytical electron microscopes.
The Sakai et al patent describes a method where electrically neutral active species are formed outside of the chamber in a plasma discharging gas and passed through a selection device which stops the ions and electrons from entering the chamber. The plasma described is generated in a microwave cavity and then carried to the chamber in a tube. This method destroys some of the active neutral species by wall collisions on the way to the chamber whereby its effectiveness is reduced. The preferred embodiment of the Sakai et al method also uses a oxygen and CF4 gas mixture as the reactive gas. This mixture or pure oxygen can cause explosive conditions in vacuum pumps using conventional hydrocarbon pump oil. Electron microscopes commonly use this oil and its replacement involves an expensive rebuild and cleaning of the pumps to accept non-reactive fluorocarbon oil. For oxygen to be used as a cleaning gas in electron microscope it must be diluted with a inert gas such the noble gases or nitrogen to avoid this explosion hazard.
A device for cleaning electron microscope stages and specimens is described in U.S. Pat. No. 5,510,624 (Zaluzec) for analytical electron microscopes. That apparatus uses an plasma generating chamber and an airlock to allow the specimen and stages to be placed into the plasma chamber for cleaning. It may be connected with the analytical chamber of the analytical electron microscope. A refinement of this device is described in U.S. Pat. No. 5,633,503 (Fichone). This device cleans by the use of sputter etching and fragmentation when argon is used as the reactive gas. When oxygen is used reactive ion etching adds to the cleaning effect. Sputter etching and reactive ion etching are anisotropic processes. This has two effects, cleaning is not evenly distributed but is concentrated to where there is ion bombardment of surfaces, and the ion bombardment may cause unwanted of sputter deposition and etching within the analytical chamber resulting in damage to the instrument and specimen.
The previous related invention of R. Vane (U.S. patent application Ser. No. 09/054,749) for glow discharge cleaning within the specimen chamber of the electron microscope using a plasma generator in-situ within the specimen chamber was found to have several disadvantages in practice as follows:
a) The use of argon and other noble gases is not good for the ion pumps used on the electron columns of many electron microscopes. Noble gases are not pumped well by dipole ion pumps. PA1 b) Generating a free plasma inside the specimen chamber may subject the walls and surfaces within the chamber to electron and ion bombardment which may cause damage. PA1 c) Some ion species in some gas-plasmas will polymerize the surface hydrocarbons to inhibit further oxidation and cleaning.
It is an object of the present invention to provide an improved method for cleaning the specimen chamber, specimen stage and a specimen in the vacuum system of an electron microscope or similar analytical instrument using an electron beam such as a scanning electron microprobe instrument. It is another object of the present invention to provide an improved cleaning method and apparatus that uses air passed through a low powered glow-discharge or gas-plasma as source of oxygen radicals to oxidize the hydrocarbon contaminants in the specimen chamber and convert them to easily pumped gases. It is another object of the present invention to minimize contact and ion bombardment of the specimen and specimen stage by the electrically active gas plasma containing ions and electrons that may cause damage. It is another object of the present invention to provide a cleaning system that is small and that can be mounted on a standard chamber port of the electron microscope without mechanical interference from other devices and parts of the electron microscope. This improvement results in a cleaning system that is faster and cleans the specimen chamber, stage, and specimen of the analytical instrument better than previous arrangements. The result of a cleaner specimen, specimen chamber and stage is that the deposition of hydrocarbon polymer on the scanned area is reduced or eliminated resulting in better measurements. Another result of cleaner specimen chambers is that the condensation and adsorption of hydrocarbons on detector windows is reduced which allows the passage of more low energy x-rays and electrons through these windows.